Region Selection by Markers may have a Bug in it

toodamnhip · Post by **toodamnhip** » Wed Feb 18, 2015 1:27 am

I have noticed a strange little quirk of DP.

When selecting a region of data in a song by clicking on Marker Icons in various windows, SOME data doesn’t get selected.

For example:
Bar 50-60, you click on the Marker at Bar 50, and, the data in that region selects. But, some of the data EXACTLY at bar 50 does NOT get selected. This might be a MIDI note or a piece of vol automation,...whatever. If gets “left behind” or doesn’t select...
Keep in mind the marker location point is EXACTLY at bar 50 so no data should be missing in the region selected, yet, again, it misses data right at the marker. (bar 50, beat 1).
However, if you MANUALLY drag from bar 50-60 in that same Seq editor window, the MIDI or another data EXACTLY at bar 50 DOES get selected.
It’s as if there is a bug when selecting a region by markers.

Can anyone verify this? By the way, it is not a 100% constant issue. Sometime the data at the marker does select. It’s a bit strange and , to me, this makes selecting regions by markers something one has to be cautious about, double checking that you got ALL your data at the front end of the selection.

Shooshie · Post by **Shooshie** » Wed Feb 18, 2015 5:47 am

toodamnhip wrote:I have noticed a strange little quirk of DP.

When selecting a region of data in a song by clicking on Marker Icons in various windows, SOME data doesn’t get selected.

For example:
Bar 50-60, you click on the Marker at Bar 50, and, the data in that region selects. But, some of the data EXACTLY at bar 50 does NOT get selected. This might be a MIDI note or a piece of vol automation,...whatever. If gets “left behind” or doesn’t select...
Keep in mind the marker location point is EXACTLY at bar 50 so no data should be missing in the region selected, yet, again, it misses data right at the marker. (bar 50, beat 1).
However, if you MANUALLY drag from bar 50-60 in that same Seq editor window, the MIDI or another data EXACTLY at bar 50 DOES get selected.
It’s as if there is a bug when selecting a region by markers.

Can anyone verify this? By the way, it is not a 100% constant issue. Sometime the data at the marker does select. It’s a bit strange and , to me, this makes selecting regions by markers something one has to be cautious about, double checking that you got ALL your data at the front end of the selection.

My guess, Dave, is that it selects "non-inclusive." I haven't tested that hypothesis in a while, but it seems like I once arrived at that conclusion. So, it would select up to 4/4/479, but not 5/1/000.

There's a quick fix for selections made like that, but it requires a key command that you may have to set up. Figure how often you'll use it and choose your keys carefully for repeated usage. Look in the Tips Sheet page 2, in the post about using selections, and you'll find commands for extending range forward and backward "a little bit." That is, you can capture any stragglers who want to avoid being selected.

Here are the pertinent illustrations, based on the way I have those commands set up:

Setting up the Commands Window:

Using the keys:

Setup: (1st illustration) and using it:asdws (illustrations 5 & 6).

I hope that helps... a little....

Shoosh

frankf · Post by **frankf** » Wed Feb 18, 2015 8:39 am

You may have Smart Selections turned on. Last item in the Edit menu

toodamnhip · Post by **toodamnhip** » Wed Feb 18, 2015 12:02 pm

frankf wrote:You may have Smart Selections turned on. Last item in the Edit menu

"Smart selections" could be the cause?
Wow....
And Shoosh, what you wrote still may apply but the problem I am having is with the front note not being selected. I will check out what you wrote though thanks

Shooshie · Post by **Shooshie** » Wed Feb 18, 2015 5:39 pm

toodamnhip wrote:
frankf wrote:You may have Smart Selections turned on. Last item in the Edit menu

"Smart selections" could be the cause?
Wow....
And Shoosh, what you wrote still may apply but the problem I am having is with the front note not being selected. I will check out what you wrote though thanks

Oh, the front note? You mean, 1/1/000? I haven't seen that. I wonder what's causing it. What edit window are you using? [never mind; I see you're using the Sequence Editor]

Shooshie

Shooshie · Post by **Shooshie** » Wed Feb 18, 2015 6:29 pm

Just checked it out. Indeed, I'm seeing it. The reason I do not see it normally is because I just assume that my notes are not on the beat exactly. I always use the methods I posted in my first post up above to rein in stray notes and controllers when making selections.

Smart Selections was off. When I turned on Smart Selections, it brought those stray notes into the selection when merely clicking on the marker. That's working as it should.

BUT... It's not exactly a bug. Let me explain.

I zoomed out to maximum resolution. As I did, this note that I thought was right on the beat (10.1.000) started moving left. It actually was located slightly BEFORE the beat, but not enough to show up in the counter. The counter only resolves to 480 tics. The edit windows resolve to sample accuracy, or close to it. So, when rounding to 480 tics, the edit window places the note at beat 10/1/000, when actually it's at 9/4/479.96, or something along those lines.

So, Selection sees that, but we don't. You have a few options:
1) quantize notes around the boundaries of markers,
2) use the method I described above to rein in ragged edges.
3) use smart selections.

You could also change the tic resolution to some absurdly high number in the Time Formats dialog. Not sure if this would display the accurate number or not.

Shooshie

monkey man · Post by **monkey man** » Thu Feb 19, 2015 12:52 am

Yeah, I figured the data must've been close to the beat, but slightly before it, Shoosh.

FWIW, I zoom in a little (usually this is necessary in the TO), and extend the selection a tad holding down the command key for precision. Or, I start with said key on the left and drag to the right. Seems quick enough to me.

frankf · Post by **frankf** » Thu Feb 19, 2015 2:36 am

I'll throw something else out: are the markers exactly where you think they are? How did you place them? Using the counter to locate then a key command to place them? I'm usually working in a finer time unit, frames and sub frames, than 480 ticks (bars and beats). There's also samples available as a time unit. Try switching to one of the finer time units and look at the Event Information window and see if your marker and note are at the same exact location.

Frank Ferrucci

toodamnhip · Post by **toodamnhip** » Thu Feb 19, 2015 2:43 am

That's some good detective work guys. I hate to burst your bubble but, the data was quantized. so unless Digital performer quantized things wrong, this must be a bug.

frankf · Post by **frankf** » Thu Feb 19, 2015 2:53 am

toodamnhip wrote:That's some good detective work guys. I hate to burst your bubble but, the data was quantized. so unless Digital performer quantized things wrong, this must be a bug.

Raises an interesting question, can markers be quantized? I've never even tried. I was talking about the marker locations possibly being off by some slight amount. I'm going to test this myself later.

Frank Ferrucci

toodamnhip · Post by **toodamnhip** » Thu Feb 19, 2015 3:16 am

frankf wrote:
toodamnhip wrote:That's some good detective work guys. I hate to burst your bubble but, the data was quantized. so unless Digital performer quantized things wrong, this must be a bug.
Raises an interesting question, can markers be quantized? I've never even tried. I was talking about the marker locations possibly being off by some slight amount. I'm going to test this myself later.

Frank Ferrucci

Markers DO quantize depending upon whether or not you are in grid mode when you place them

frankf · Post by **frankf** » Thu Feb 19, 2015 4:14 am

That makes sense, though I do not place markers by dragging onto the grid. Anyway I just tested your original "bug" report here by adding 2 markers interior to a piece using locate, then Cmd-M at 19-1-0 and 30-1-0, locating notes to exactly 19-1-0 by 1) typing 19-1-0 as start time in Event Information Window and 2) using Quantize with 100% Quantize, then clicking on the marker at 19-1-0 to select all between 19-1-0 and 30-1-0. I then pasted elsewhere on the down beat of a bar. Smart Selection being off. All notes were copied and pasted as expected. I repeated this a few of times without problem.

Frank Ferrucci

Shooshie · Post by **Shooshie** » Thu Feb 19, 2015 6:36 am

Yeah, I did the same. When the notes were truly on the beat, and not even a fraction of a tic before it, they always got picked up by the Marker selection.

Are you 100% certain that the notes that were missed were not slightly before the beat?

On the other hand, I wonder if there are certain cases in which Quantize may be placing things off by a few samples. I haven't seen it, but I don't doubt that toodamnhip is seeing something, and if so, there's got to be a reason.

Shooshie

Shooshie · Post by **Shooshie** » Thu Feb 19, 2015 6:45 am

toodamnhip wrote:I hate to burst your bubble but, the data was quantized.

Just for grins, could you quantize those notes again, and see if it still happens?

Also, zoom in as far as you can and see if the markers are exactly where they are supposed to be.

frankf · Post by **frankf** » Thu Feb 19, 2015 12:27 pm

And, all the Quantize options, humanize, offset, strength, etc. must be off to get a hard quantize. Though I agree. If he's getting this behavior something is going on, a setting he has or his system. I can't think of anything else at this point.

MOTUnation.com